Novel treatments are warranted for peritoneal surface malignancies (including colon, ovarian, appendiceal, and mesothelioma). We have a comprehensive program for the clinical management of patients with peritoneal surface malignancies and are exploring new therapeutic options. We also have a long standing interest in poxviruses for oncolytic viral therapy, and we previously developed a tumor-selective, replicating oncolytic vaccinia virus for clinical use (vvDD). While vvDD has demonstrated success as a direct injection into patients' tumors, it is limited by the patient's premature immune mediated clearance of the virus. We propose in this project to enhance this vaccinia virus for the treatment of peritoneal tumors, addressing the limitation of premature immune clearance of the virus. We recently discovered that the expression of chemokines, CCL5 or CCL19, leads to prolonged, selective replication in the tumor microenvironment, however, in the normal tissues the chemokine-expressing virus is cleared rapidly. We hypothesize that the CKs (CCL5 or CCL19) secreted from poxvirus-infected cancer cells in the tumor have attracted from circulation a large amount of naive leukocytes including monocytes and lymphocytes. These cells in the tumor microenvironment initiate and sustain strong type 2 immune responses which enable the CCL5 (or CCL19)-expressing virus to persist in the tumor. Pre-existing anti-poxviral immunity also affects the efficacy of vvDD. We have recently demonstrated that an A34R mutation enhances the release of an enveloped form (EEV) of the virus, which may not be recognized by neutralizing antibodies. The A34R mutation has the additional advantage of releasing large quantities of virus from the cell, enhancing the spread of the virus to distant sites. This is especially important in the setting of peritoneal spread, where it is common to have thousands of individual tumor deposits. Thus, we hypothesize that our tumor-selective vvDD, when engineered with an A34R mutation and the appropriate chemokine expression, will result in improved treatment of peritoneal surface malignancies, and overcome important limitations encountered in our clinical trial. Our aims are: (1) Characterize the extent and mechanism of prolonged vvDD replication in vivo when expressing CCL5 and/or CCL19 chemokines, (2) Define the mechanism of prolonged viral replication using in vitro modeling, and gene knockout mouse models, and (3) Arming vvDD with the ability to evade the pre-exisiting anti-poxviral immunity and spread widely throughout the peritoneal cavity, by mutating the A34R gene. With these aims we are confident that we will overcome some of the obstacles to successful viral therapy and be able to move forward with a new clinical trial for patients with peritoneal surface malignancies. The information obtained regarding the immune consequences of chemokine expression in the tumor microenvironment will be important for all viral therapies. The significance of the A34R deletion with enhanced release of the vvDD will be important for all poxvirus clinical applications.